Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

Re: factory-made roof trusses

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Let me add some of my fun experiences with factory roof trusses.  I have
questioned a few suppliers and their network of technical information.  Here
are the results of several different suppliers:

1.   The plate connections in some cases are "tested" values.  You could not
duplicate their strength with numerical calculations.  The values are ONLY
good if the plate is "pressed" into place in a factory setting.  For this
reason, most companies have special rules for handling trusses.  They do not
permit you to handle the truss flat and therefor bow the truss down.  This
weakens the plate connection.  Just about every job I have visited has them
handling the trusses flat.  They are delivered bowed in a lot of cases.  So
field handling did not match requirements.  If they "pop" a plate loose in the
field, they just hammer it back in place.

2.  Truss designs with large open rooms in the web.   These non-triangular
rooms (and 5 sides in some cases) have not been checked for unbalanced loading
if the code does not require an unbalanced case.  Their computer model placed
a uniform load over all rafters and the truss checked out.  When I required it
to take a RLL one side only, it failed out in the model.  The model was stable
in the computer only, not in the real world.  The computer can stack marbles 3
miles high, if it is programmed right.  I was basically questioning the
stability of the truss as a whole with this large open room and ALL pinned
joints.  I felt RLL one side only was a reasonable requirement since that is
how this type of roof gets built, one side at a time.

3.  Some programs I investigated were allowing you to put in a distributed
load along the chords, but was immediately converting it to  a joint load and
applying it to the joints.  No analysis within the model was made of the beam-
nature of the chords. When asked why they do not analyze it as it actually is
used, the answer was "Everyone knows trusses are loaded at the
joints"....scary comment

4.  The center of the bottom chord on a truss with a triangular section in the
middle. I find that bottom chord/joist with a standard gangnail plate, and no
real ability to resist moment.  But since the program converted all
distributed  loads to  joint loads, it did not realize the problem.  Add to
this, that this is one of the main joints that "pops' loose when handling the
truss flat and I get a little leary sometimes walking on truss bottom chords
in the middle.  I probably induce more LL than most others on this List.

As I stated earlier, this is several items I have run into from several
different truss programs, and not all from one program only.

Good Luck